Marion County Oregon - Prescribed Burn Plan

Bonesteele Park

Prescribed Burn Plan

Home > Public Works > Parks > Natural Heritage Parks Program > Bonesteele Park > Prescribed Burn Plan

Fuel load along a gradient at Bonesteele Park, Marion County, Oregon

Sara Coffey
Science Collaberative Research Program
Willamette University
Environmental and Earth Sciences
Summer 1999

Abstract:
This study examines fuel load at Bonesteele Park, Marion County, OR. Data were obtained along several transects from edge to interior of the forest. The total fuel load calculated ranged between 1.65 to 7.25 tons/ha. The fuel load did not vary along the gradient from the edge to the interior of the forest. Bonesteele Park may not be large enough to contain a fuel gradient. The low fuel load maybe due to logging of trees that would have otherwise fallen down and created fuel and the rate of decay. Because the fuel load is low a fire would not sustain at Bonesteele Park during July, the time of the study. It would be more effective for burning to occur in the fall when there is litter to sustain a fire. Ongoing monitoring of fuel loads is advised as the park undergoes vegetation change through restoration efforts.
Introduction:
When Europeans first arrived on the Willamette Valley their eyes focused on a land extremely different from what we see today. It is believed that grasses dominated the Willamette Valley before European settlement. Native Americans depended on lightning and accidental fires to manage the landscape. Before European settlement, landscape dynamics in Pacific Northwest was driven primarily by the patterns of wildfire (Wallin 1996). The pre-settlement disturbance regime included wildfires that, under extreme conditions, could consume over 1000 Km of forest in a single event (Agee 1993).
Before European settlement the Willamette Valley is believed to be an area dominated by grasslands and oak savanna. The land soon began to change once Europeans settled upon it. The large influx of European settlement occurred in the 1840s. Europeans indirectly influenced the fire regime by reducing Native American populations, which reduced the frequency of ignitions by Native Americans. Fire suppression began in 1910 (Wallin 1996). There was a substantial reduction in the area of natural vegetation since European settlement because much of the land base has been usurped for buildings, agriculture, or intensive, short-rotation industrial forestry (Wallin 1996). With out burning, Oregon White Oak (Quercus garryana) and tall grass stands were gradually replaced by the better adapted Douglas fir (Pseudotsuga menziesii) bigleaf maple (Acer maerophyllum) and grand fir (Abies grandis) woodlands removing the tall grass prairies dominate presence in the valley (Kimmerling and Jackson, 1985:61 as cited from Kenneth Duncan 1999).

At Bonesteele Park in Marion County, Oregon, land managers are interested in encouraging an open oak savanna type edge between restored prairie and upland forest. Other restoration projects have shown that prescribed fire is a useful management tool for restoring vegetation. In order to implement prescribed burning, one must characterize the fuels, since they drive fire. These characteristics include fuel load and seasonality of fuels. Accurate measurements of surface fuel loadings and consumption during prescribed fires is important for understanding the effects of fire on vegetation (Finney 1993).

Research Objectives:
Marion County Public Works Department is conducting a project, which focuses on restoring the vegetation of its 30 acre Bonesteele Park to pre-European conditions. The goal of the parks department is to convert the upland pasture (20 acres) into an upland prairie and convert the edges of the forest to an oak savanna. They plan to remove the species that are non-native to the site and replace this vegetation with the vegetation that flourished before Europeans colonized.
In this study I will focus on the fuel load in the forested area. The main question that I intend to answer is: Are fuel conditions such that prescribed fire can be used along the prairie/forest ecotone? I answer this question by measuring the fuel load differences between the edge and the interior of the forest. The purpose of examining the fuel load is to help land managers practice fuel management and plan for prescribed fire (Brown 1974). The identification of fuels can give landowners an opportunity to plan and understand the potential effect on vegetation of certain fuel levels. It also allows managers to plan ways to reduce the hazardous effects that may result if wild fire occurs.

In order to determine if the fuel conditions at Bonesteele Park are such that prescribed fire can be used along the prairie/forest ecotone I will test for significant differences among fuel loads, along a transect from prairie to interior. I hypothesize that the fuel load will increase from the edge of the forest throughout the gradient into the interior because of the finer fuels at the edge and the larger fuels in the interior.

Site Description:
Bonesteele Park is a 30 acre upland site located southeast of Salem, Oregon (Figure 1). 20 acres are former farming field, now in grass Festuca ovina var Duriuscula (hard fescue) and a 10 acre mature forest dominated by Psuedotsugs manziesii (Douglas Fir), Acer macrophyllum (Bigleaf Maple), and Corylus corinuta (Hazelnut).
The soils at Bonesteele Park are Nekia silty clay loam (Marion County Soil Survey 1972 as cited in Ridgeline Resource Planning). The site has a slope between 5-20% as determined from the U.S.G.S. topographic mapping. Bonesteele slopes form north to south with an overall southern exposure. Annual precipitation is measured in nearby Salem (39.24 in/yr) and Stayton (51.48 in/yr) (Taylor & Bartlett 1993 as cited in Ridgeline Resource Planning.

Field Methods:
We established four 90m North-South transects spaced 35m apart (figure 1). A 200m2 plot was placed every 20m along each transect for a total of 5 plots per transect (20 plots total). Each plot represented a different vegetation type as follows: prairie, forest edge, transition between edge and interior forest, and interior forest. Downed woody fuels were measured following Brown (1974). I took two fuel samples in each plot, 2.5 meters from both the north and south sides, I ran a tape 15m in a random direction. There are four different types of fuel: one hour (0-0.64cm diameter), ten hour (0.65-2.54cm diameter), one hundred hour (2.55-7.62cm diameter), and one thousand hour (>7.62cm diameter). One-hour fuels were counted between 0-2m along the tape. Ten hour fuels were counted between 0-3m. One hundred-hour fuels were counted between 0-5m. One thousand-hour fuels were counted the whole 15m of the tape. The one thousand-hour fuels were divided in two categories, sound or rotten, and the diameter at breast height (DBH) was recorded for each. Then measurements of fuel, litter, and duff were taken at 0m, 3m, 6m. This was done by using a trowel to display a soil profile and measuring the depth of each. Finally, because bulk density of this litter type is not known, a known volume of litter was collected in each plot to calculate bulk density. A 20cmX20cm (400cm2) square was placed on the ground under the 7.5-meter mark and was outlined. Then all the litter under that square was collected and this depth noted to obtain a volume. Then, the litter samples were weighed, then oven dried at 100 degrees Celsius for 24 hours, and weighed again to obtain the bulk density, which was used to calculate litter fuel load. The raw fuel data were converted to fuel load (tons/ha) following Brown (1974).
Lab Methods:
Because the data were not normal non-parametric statistics tests, Kruskal-Wallis and Mann Whitney were used to test for differences in fuel loads between plots at p<0.05.
Results:
The results indicate some variation in fuel load among the different vegetation types along the transect. The results from the Kruskal-Wallis test (p<0.05) showed there were significant differences in 1hr, 10hr, litter, and total fuel loads (table 1). All of these values are adjusted for ties.
Pairwise comparisons between the five plot types were run using the Mann-Whitney test on the fuel categories where significant differences existed (Table 2). Litter load differed significantly between the prairie and the edge plots (p=0.0265), the prairie and interior plots (p=0.0211), and the prairie and the deep interior (p=0.0265).

One hour fuel loads differed significantly between the prairie and the transition (p=0.0265); between the prairie and the interior (p=0.0265); between the prairie and the deep interior (p=0.0265); and between the middle edge and the deep interior plots (p=0.0304).

Ten hour fuel loads differed significantly between the prairie and the transition (p=0.0265); between the prairie and the interior (p=0.0265); between the prairie and the deep interior (p=0.0265).

Total fuel loads differed significantly between the prairie and the edge (p=0.0265); between the prairie and the transition (p=0.0265); between the prairie and the interior (p=0.0265); and between the prairie and deep interior (p=0.0265).

Total fuel loads range from 1.65 to 7.25 tons/ha.

Discussion:
There are many factors contributing to the low fuel load at Bonesteele Park. Due to restoration efforts the prairie had recently been herbicided and plowed therefore at the time of this census no fuels were present in the prairie plots. As a result, fuel loads in the prairie were significantly different from the transition, interior, and deep interior plots for one-hour and ten-hour fuels. The data for one-hour fuel is higher in the transition plots because these plots intersected a large pile of brush left as part of the restoration work. The edge was not significantly different from the prairie for one-hour and ten-hour fuels because of the staggered plow line separating the two plots. Sometimes a portion of the prairie was contained in the edge plots. Due to the low fuel load throughout the gradient a fire would not be able to persist. This study was conducted in July and there are very little amounts of litter at this time. Litter is what carries a fire. Therefore, prescribed burning would be most effective when litter loading is at its highest, in the fall.
Fuels at Bonesteele Park forest can be categorized in the fuel model 9, which is based on the type and amount of fuel present. Fuel model 9 is a mixed hardwood coniferous forest. Knowing the fuel model enables land managers to predict fire behavior and to rate fire danger. Based on these results, the low fuel loads and fuel model 9 characterization indicate that the best conditions for prescribed burning most likely occur in the early fall while it is still dry, but after leaf fall has occurred. Summer burning would not be advisable due to patchy, low fuel loads.

Conclusion:
Fuel loads do not vary along a gradient at Bonesteele Park. The most effective time to burn is in the fall at the prairie/forest edge. Continued monitoring of the fuel load is important during restoration because of vegetation changes, such as removal of non-native vegetation, will change fuel loads.
References:

Agee, James. 1993. Fire Ecology of Pacific Northwest Forests. Washington D.C.: Island Press.

Brown, James. 1974. Handbook for Inventorying Downed Woody Material. USDA Forest Service.

Finney, M.A. 1993. Fuel loading, bulk density, and depth of forest floor in coast redwood stands. Forest Science 39:617-622.

Ridgeline Resource Planning. 1999. Bonesteele Park Upland Prairie Restoration. [Online] Available: http;//www.org/mcpark/restore.html.

Wallin, D.O., Swanson, I.J., Marks, B., Cissel, J.H., and J. Kerlis. 1996. Comparison of managed and pre-settlement landscape dynamics in forests of the Pacific Northwest, U.S.A. Forest Ecology and Management 85: 291-309.

Directory | Services | Employment | Volunteer
E-mail comments or suggestions to webmaster@co.marion.or.us